This research involves three general components: prediction during the prenatal and perinatal period of families who will experience disorders of parenting; understanding of the factors and processes occurring prior to the manifestations of these disorders of parenting; and prevention of disorders of parenting through early and extended contact between parent and newborn following delivery. A low-income, southern, urban population delivering at one county hospital is studied. All 1,700 pregnant women entering the prenatal clinic over a 17-month period were screend prenatally for risk for subsequent parenting problems utilizing an interview developed by the investigators. 273 High Risk women, 274 Randomly Sample women, and a third small identified group of families known to have neglected or abused a child are followed prospectively. All subjects in the High Risk and Random Sample groups are studied until the target child, unborn at the time of screening, reaches 18 months of age. The following data are collected: incidence of parenting disorders; maternal attitudes toward childrearing; Brazelton Neonatal Assessments; mother-infant interaction observations; maternal perceptions of infant's temperament; performance of the target child on the Bayley Scales. These data are analyzed in order to derive a predictive equation of factors which result in disorders of parenting and in order to develop an understanding of the interactional and environmental processes which yield child maltreatment. The identified group is studied in order to assess the validity of data obtained retrospectively. The prevention experiment involves 200 of the subjects who are in either the High Risk or the Random Sample group, plus 1,000 additional women in neither group, who were randomly assigned at the time of delivery to experience routine hospital-allowed contact with their newborns or to experience extended contact with their newborns.